{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T04:43:46Z","timestamp":1790743426541,"version":"4.1.0"},"reference-count":30,"publisher":"Oxford University Press (OUP)","issue":"3","license":[{"start":{"date-parts":[[2021,12,15]],"date-time":"2021-12-15T00:00:00Z","timestamp":1639526400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001505","name":"Health Research Council of New Zealand","doi-asserted-by":"publisher","award":["11\/800"],"award-info":[{"award-number":["11\/800"]}],"id":[{"id":"10.13039\/501100001505","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001505","name":"Health Research Council of New Zealand","doi-asserted-by":"publisher","award":["14\/010"],"award-info":[{"award-number":["14\/010"]}],"id":[{"id":"10.13039\/501100001505","id-type":"DOI","asserted-by":"publisher"}]},{"name":"New Zealand Health Research Council Clinical Research Training Fellowship"},{"DOI":"10.13039\/501100007281","name":"National Drug and Alcohol Research Centre","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100007281","id-type":"DOI","asserted-by":"crossref"}]},{"name":"University of New South Wales Scientia PhD Scholarships"},{"name":"New Zealand Heart Foundation Hynds Senior Fellowship"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2022,6,13]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec><jats:title>Background<\/jats:title><jats:p>Machine learning-based risk prediction models may outperform traditional statistical models in large datasets with many variables, by identifying both novel predictors and the complex interactions between them. This study compared deep learning extensions of survival analysis models with Cox proportional hazards models for predicting cardiovascular disease (CVD) risk in national health administrative datasets.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>Using individual person linkage of administrative datasets, we constructed a cohort of all New Zealanders aged 30\u201374\u2009who interacted with public health services during 2012. After excluding people with prior CVD, we developed sex-specific deep learning and Cox proportional hazards models to estimate the risk of CVD events within 5 years. Models were compared based on the proportion of explained variance, model calibration and discrimination, and hazard ratios for predictor variables.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>First CVD events occurred in 61\u2009927 of 2\u2009164\u2009872 people. Within the reference group, the largest hazard ratios estimated by the deep learning models were for tobacco use in women (2.04, 95% CI: 1.99, 2.10) and chronic obstructive pulmonary disease with acute lower respiratory infection in men (1.56, 95% CI: 1.50, 1.62). Other identified predictors (e.g. hypertension, chest pain, diabetes) aligned with current knowledge about CVD risk factors. Deep learning outperformed Cox proportional hazards models on the basis of proportion of explained variance (R2: 0.468 vs 0.425 in women and 0.383 vs 0.348 in men), calibration and discrimination (all P\u2009&amp;lt;0.0001).<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusions<\/jats:title><jats:p>Deep learning extensions of survival analysis models can be applied to large health administrative datasets to derive interpretable CVD risk prediction equations that are more accurate than traditional Cox proportional hazards models.<\/jats:p><\/jats:sec>","DOI":"10.1093\/ije\/dyab258","type":"journal-article","created":{"date-parts":[[2021,11,26]],"date-time":"2021-11-26T12:08:48Z","timestamp":1637928528000},"page":"931-944","source":"Crossref","is-referenced-by-count":50,"title":["Predicting cardiovascular risk from national administrative databases using a combined survival analysis and deep learning approach"],"prefix":"10.1093","volume":"51","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-5919-372X","authenticated-orcid":false,"given":"Sebastiano","family":"Barbieri","sequence":"first","affiliation":[{"name":"Centre for Big Data Research in Health, University of New South Wales , Sydney, NSW, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Suneela","family":"Mehta","sequence":"additional","affiliation":[{"name":"Section of Epidemiology and Biostatistics, University of Auckland , Auckland, New Zealand"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Billy","family":"Wu","sequence":"additional","affiliation":[{"name":"Section of Epidemiology and Biostatistics, University of Auckland , Auckland, New Zealand"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chrianna","family":"Bharat","sequence":"additional","affiliation":[{"name":"National Drug and Alcohol Research Centre, University of New South Wales , Sydney, NSW, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Katrina","family":"Poppe","sequence":"additional","affiliation":[{"name":"Section of Epidemiology and Biostatistics, University of Auckland , Auckland, New Zealand"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0390-661X","authenticated-orcid":false,"given":"Louisa","family":"Jorm","sequence":"additional","affiliation":[{"name":"Centre for Big Data Research in Health, University of New South Wales , Sydney, NSW, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5914-6934","authenticated-orcid":false,"given":"Rod","family":"Jackson","sequence":"additional","affiliation":[{"name":"Section of Epidemiology and Biostatistics, University of Auckland , Auckland, New Zealand"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2021,12,15]]},"reference":[{"key":"2022061309574443800_dyab258-B1","doi-asserted-by":"crossref","first-page":"i2416","DOI":"10.1136\/bmj.i2416","article-title":"Prediction models for cardiovascular disease risk in the general population: systematic review","volume":"353","author":"Damen","year":"2016","journal-title":"BMJ"},{"key":"2022061309574443800_dyab258-B2","doi-asserted-by":"crossref","first-page":"e008717","DOI":"10.1136\/bmjopen-2015-008717","article-title":"Impact of provision of cardiovascular disease risk estimates to healthcare professionals and patients: a systematic review","volume":"5","author":"Usher-Smith","year":"2015","journal-title":"BMJ Open"},{"key":"2022061309574443800_dyab258-B3","doi-asserted-by":"crossref","first-page":"859","DOI":"10.1136\/jech-2012-200971","article-title":"Predictive risk algorithms in a population setting: an overview","volume":"66","author":"Manuel","year":"2012","journal-title":"J Epidemiol Community Health"},{"key":"2022061309574443800_dyab258-B4","doi-asserted-by":"crossref","first-page":"380","DOI":"10.1093\/ije\/dyp373","article-title":"Commentary: assessing population (baseline) risk is a cornerstone of population health planning\u2014looking forward to address new challenges","volume":"39","author":"Manuel","year":"2010","journal-title":"Int J Epidemiol"},{"key":"2022061309574443800_dyab258-B5","doi-asserted-by":"crossref","first-page":"1571","DOI":"10.1093\/ije\/dyy137","article-title":"Development and validation of alternative cardiovascular risk prediction equations for population health planning: a routine health data linkage study of 1.7 million New Zealanders","volume":"47","author":"Mehta","year":"2018","journal-title":"Int J Epidemiol"},{"key":"2022061309574443800_dyab258-B6","doi-asserted-by":"crossref","first-page":"187","DOI":"10.1111\/j.2517-6161.1972.tb00899.x","article-title":"Regression models and life-tables","volume":"34","author":"Cox","year":"1972","journal-title":"J R Stat Soc Ser B Methodol"},{"key":"2022061309574443800_dyab258-B7","doi-asserted-by":"crossref","first-page":"233","DOI":"10.1038\/nmeth.4642","article-title":"Points of significance: statistics versus machine learning","volume":"15","author":"Bzdok","year":"2018","journal-title":"Nat Methods"},{"key":"2022061309574443800_dyab258-B8","first-page":"1","article-title":"Time-to-event prediction with neural networks and Cox regression","volume":"20","author":"Kvamme","year":"2019","journal-title":"J Mach Learn Res"},{"key":"2022061309574443800_dyab258-B9","volume-title":"National Health Index Data Dictionary (Version 5.3) [Internet]","year":"2009"},{"key":"2022061309574443800_dyab258-B10","volume-title":"Cardiovascular Disease Risk Assessment. Updated 2013 (New Zealand Primary Care Handbook 2012).","year":"2013"},{"key":"2022061309574443800_dyab258-B11","doi-asserted-by":"crossref","first-page":"133","DOI":"10.2147\/CLEP.S138100","article-title":"Cardiovascular medication changes over 5 years in a national data linkage study: implications for risk prediction models","volume":"10","author":"Mehta","year":"2018","journal-title":"Clin Epidemiol"},{"key":"2022061309574443800_dyab258-B12","doi-asserted-by":"crossref","first-page":"1885","DOI":"10.1162\/089976699300016007","article-title":"Combined 5 \u00d7 2 CV F test for comparing supervised classification learning algorithms","volume":"11","author":"Alpaydm","year":"1999","journal-title":"Neural Comput"},{"key":"2022061309574443800_dyab258-B13","article-title":"Adam: A method for stochastic optimization","author":"Kingma"},{"key":"2022061309574443800_dyab258-B14","article-title":"Deep ensembles: a loss landscape perspective","author":"Fort","year":"2019"},{"key":"2022061309574443800_dyab258-B15","first-page":"216","article-title":"Discussion of the paper by D. R. Cox","volume":"34","author":"Breslow","year":"1972","journal-title":"J R Stat Soc B"},{"key":"2022061309574443800_dyab258-B16","doi-asserted-by":"crossref","first-page":"723","DOI":"10.1002\/sim.1621","article-title":"A new measure of prognostic separation in survival data","volume":"23","author":"Royston","year":"2004","journal-title":"Stat Med"},{"key":"2022061309574443800_dyab258-B17","doi-asserted-by":"crossref","first-page":"2543","DOI":"10.1001\/jama.1982.03320430047030","article-title":"Evaluating the yield of medical tests","volume":"247","author":"Harrell","year":"1982","journal-title":"JAMA"},{"key":"2022061309574443800_dyab258-B18","doi-asserted-by":"crossref","first-page":"2529","DOI":"10.1002\/(SICI)1097-0258(19990915\/30)18:17\/18<2529::AID-SIM274>3.0.CO;2-5","article-title":"Assessment and comparison of prognostic classification schemes for survival data","volume":"18","author":"Graf","year":"1999","journal-title":"Stat Med"},{"key":"2022061309574443800_dyab258-B19","doi-asserted-by":"crossref","first-page":"26","DOI":"10.1080\/00031305.1985.10479382","article-title":"Separating the Brier score into calibration and refinement components: a graphical exposition","volume":"39","author":"Blattenberger","year":"1985","journal-title":"Am Stat"},{"key":"2022061309574443800_dyab258-B20","doi-asserted-by":"crossref","first-page":"1092","DOI":"10.1161\/CIRCRESAHA.117.311312","article-title":"Cardiovascular event prediction by machine learning: the multi-ethnic study of atherosclerosis","volume":"121","author":"Ambale-Venkatesh","year":"2017","journal-title":"Circ Res"},{"key":"2022061309574443800_dyab258-B21","doi-asserted-by":"crossref","first-page":"e0213653","DOI":"10.1371\/journal.pone.0213653","article-title":"Cardiovascular disease risk prediction using automated machine learning: a prospective study of 423,604 UK Biobank participants","volume":"14","author":"Alaa","year":"2019","journal-title":"PLoS One"},{"key":"2022061309574443800_dyab258-B22","doi-asserted-by":"crossref","first-page":"e0174944","DOI":"10.1371\/journal.pone.0174944","article-title":"Can machine-learning improve cardiovascular risk prediction using routine clinical data?","volume":"12","author":"Weng","year":"2017","journal-title":"PLoS One"},{"key":"2022061309574443800_dyab258-B23","doi-asserted-by":"crossref","first-page":"e13389","DOI":"10.1111\/ijcp.13389","article-title":"Machine learning to predict cardiovascular risk","volume":"73","author":"Quesada","year":"2019","journal-title":"Int J Clin Pract"},{"key":"2022061309574443800_dyab258-B24","doi-asserted-by":"crossref","first-page":"e17257","DOI":"10.2196\/17257","article-title":"Accurate prediction of coronary heart disease for patients with hypertension from electronic health records with big data and machine-learning methods: model development and performance evaluation","volume":"8","author":"Du","year":"2020","journal-title":"JMIR Med Inform"},{"key":"2022061309574443800_dyab258-B25","first-page":"1","article-title":"Learning from longitudinal data in electronic health record and genetic data to improve cardiovascular event prediction","volume":"9","author":"Zhao","year":"2019","journal-title":"Sci Rep"},{"key":"2022061309574443800_dyab258-B26","doi-asserted-by":"crossref","first-page":"m3919","DOI":"10.1136\/bmj.m3919","article-title":"Consistency of variety of machine learning and statistical models in predicting clinical risks of individual patients: longitudinal cohort study using cardiovascular disease as exemplar","volume":"371","author":"Li","year":"2020","journal-title":"BMJ"},{"key":"2022061309574443800_dyab258-B27","doi-asserted-by":"crossref","first-page":"25","DOI":"10.1186\/1471-2105-8-25","article-title":"Bias in random forest variable importance measures: Illustrations, sources and a solution","volume":"8","author":"Strobl","year":"2007","journal-title":"BMC Bioinformatics"},{"key":"2022061309574443800_dyab258-B28","doi-asserted-by":"crossref","first-page":"115","DOI":"10.1186\/s12874-017-0383-8","article-title":"A comparison of the conditional inference survival forest model to random survival forests based on a simulation study as well as on two applications with time-to-event data","volume":"17","author":"Nasejje","year":"2017","journal-title":"BMC Med Res Methodol"},{"key":"2022061309574443800_dyab258-B29","doi-asserted-by":"crossref","first-page":"598","DOI":"10.1007\/s12471-010-0841-y","article-title":"Gender differences in coronary heart disease","volume":"18","author":"Maas","year":"2010","journal-title":"Neth Heart J"},{"key":"2022061309574443800_dyab258-B30","doi-asserted-by":"crossref","first-page":"426","DOI":"10.1038\/s42256-020-0218-x","article-title":"When causal inference meets deep learning","volume":"2","author":"Luo","year":"2020","journal-title":"Nat Mach Intell"}],"container-title":["International Journal of Epidemiology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/ije\/article-pdf\/51\/3\/931\/44036236\/dyab258.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/ije\/article-pdf\/51\/3\/931\/44036236\/dyab258.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,9,13]],"date-time":"2024-09-13T02:16:24Z","timestamp":1726193784000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/ije\/article\/51\/3\/931\/6463074"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,12,15]]},"references-count":30,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2021,12,15]]},"published-print":{"date-parts":[[2022,6,13]]}},"URL":"https:\/\/doi.org\/10.1093\/ije\/dyab258","relation":{},"ISSN":["0300-5771","1464-3685"],"issn-type":[{"value":"0300-5771","type":"print"},{"value":"1464-3685","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2022,6,1]]},"published":{"date-parts":[[2021,12,15]]}}}